We do not yet know what effect increased access to anti-retroviral therapy (ART) will have on risk behavior in the developing world. This study examines the effect of ART on sexual risk behavior among HIV-positive individuals in Dar es Salaam, Tanzania. A quantitative survey will be administered before individuals initiate ART and again 6 months later. Following the quantitative survey, qualitative interviews will be conducted with selected participants who fall into one of three categories of risk behavior change: positive (less risky), negative (more risky), or no change. Both quantitative and qualitative components will examine sexual risk behavior as well as possible mediating factors between ART and behavior including physical health, psychological health, relational factors, treatment adherence, and treatment optimism. In addition, it is unclear who is receiving limited supplies of ART in resource-poor settings and how these people came to be tested, seek treatment, and be selected to receive ART. A secondary research aim thus seeks to explore the characteristics of people receiving ART in Tanzania. This aim will be accomplished through quantitative, qualitative, and ethnographic methods.